Proposal and Layout Optimization of a Wind-Wave Hybrid Energy System Using GPU-accelerated Differential Evolution Algorithm

Yize Wang,Zhenqing Liu,Hao Wang
DOI: https://doi.org/10.1016/j.energy.2021.121850
IF: 9
2022-01-01
Energy
Abstract:Offshore wind turbines (OWTs) are used to harvest wind energy on the seas. Their foundations must bear wind and wave loads simultaneously, resulting in greater cost than onshore wind turbines. To harvest more energy and reduce the total cost, a wind-wave hybrid energy system is proposed in this study. Oscillating wave surge converters (OWSCs) are used to harvest wave energy and protect OWTs from waves. The reduced wave impact on the wind turbine foundation reduces the total cost. Wind turbines and converters are independent; even existing wind farms can convert to hybrid farms. An analytical wave wake model for the converter is presented to optimize the layout of the devices. The wave wake of the wind turbine foundations is also examined. An innovative GPU-accelerated multi-objective differential evolution algorithm is used to globally optimize the device layout in actual environmental conditions. The optimized hybrid farm has 37.75% greater energy output and 43.65% less wave loads on the wind turbine foundations than wind-only farms. The wave wake model for the converter has an accuracy of 94.73%. The GPU-accelerated codes can run 1136 times faster than the CPU-based codes, and are available to other researchers.
What problem does this paper attempt to address?